Multi-Agent Systems (2nd Cycle) (2012/2013)

Projects

Done

Ongoing

Proposed

  • Roli
    • Evolution of Braitenberg controllers. Goal: pray-predator behaviour
    • Implementation of Braitenberg controllers in a Lego Mindstorm robot. Goal: test vehicles 2 and 3
    • Design of a subsumption architecture such that the robot is able to perform obstacle avoidance and phototaxis
    • Evolution of a neural network controller (or any network variant, such as Boolean network) for obstacle avoidance or phototaxis or both
    • Evolution of a neural network controller (or any network variant, such as Boolean network) for collective behaviour, such as gathering or flocking
    • Evolution of a neural network controller (or any network variant, such as Boolean network) for maze solving
    • Sequence learning task in BN robotics. Goal: T-maze behaviour
    • Information measures and intelligent behaviour: evolution of a neural network controller (or any network variant, such as Boolean network) such that a group of robots maximises the average mutual information or the LMC complexity or a similar information measure
    • Implement the robot program described in Labella et al., Division of labor in a group of robots inspired by ants' foraging behavior, ACM Transactions on Autonomous and Adaptive Systems, Volume 1 Issue 1, 2006, and do some statistics.
  • Omicini
    • Stigmergic coordination in JADE & TuCSoN
      ** BDI architecture upon JADE / Jason on JADE
    • Timed coordination in ReSpecT: status & implementation
    • Situated ReSpecT: status & implementation
    • Coordination in JADE
    • Speedy mouse vs. lazy mouse: ACLT on TuCSoN
      ** ACLT "demo" operations vs. tuProlog & ReSpecT: theory & implementation
    • Biochemical coordination in TuCSoN: status & implementation
    • tuProlog for ReSpecT & TuCSoN: dependency, profiling, and optimisation
    • DALI on tuProlog
      ** DCaseLP with Jess, JADE and tuProlog: state-of-the-art and possible developments
    • Exploring Uniform Primitives' Expressiveness: A catalog of Self-Organising Patterns (Modelling, designing and implementing a "catalog" of self-organising patterns/mechanisms/behaviours enabled by TuCSoN uniform primitives solely)

Previous Years' Projects

Multi-Agent Systems (2nd Cycle) (2012/2013)